Crosscut

This series of notes started with "How do we access record field? Maybe use functions." and went on to "How do functions even work?". Now my latest drafts, which are currently a few days ahead of what I already published, are getting into "How does type inference interact with traits?", which is pretty far of from where I started.

I think it's time to take a step back, recognize that I'm in for more than I bargained for, and reconsider how to continue. But this endeavor has been useful in one respect: If I had to implement records soon, I now know that I would just introduce special syntax for that. It's a good-enough placeholder, and anything else would be too complicated for now.

With that out of the way, I'm not sure what to do. Maybe dive head-first into Pandora's box and attack some of those gnarly questions that have started to claw their way out. Or maybe go with an easier topic first, while I think some more about those questions in the background. I guess we'll see tomorrow!